An earlier IndustryWeek article, “When Finance Runs the Factory,” ties together financial metrics, inventory policy and offshoring as its primary arguments in support of a conclusion that the office of finance is having a serious negative impact on the U.S. manufacturing industry. These supporting arguments, covering inventory, cycle time, marginal cost and globalization misunderstand the role of finance in informing, advising and running a manufacturing business. I’d like to briefly address each of the first three topics before concluding with a more comprehensive assessment of globalization’s root causes and effects.

Inventory

There would be no argument from finance that inventory be held to the lowest reasonable levels that still permit the enterprise to operate at peak efficiency, the operable word being “reasonable.” In fact, a better term would be optimal -- not too much, not too little. No manufacturing executive wants unnecessary capital being unproductively tied up in excess inventory, along with the concomitant additional carrying costs, but neither do they want to jeopardize operations by being an inventory miser.

Cash is king. In the 25 years I spent in manufacturing finance, my driving concern was the cash-to-cash cycle. I may have saved a little on up-front cash with a stringent inventory policy, but I also understood that a too restrictive inventory policy could cost me dearly in production interruptions and lost revenue, lengthening and negatively affecting the resulting cash-to-cash cycle. 

Inventory is a risk management tool. Inventory is essentially an insurance policy against the variability of the supply chain, production processes and the market, with inventory optimization as the appropriate analytical approach to balancing risk with return. Inventory optimization operates with two inputs -- a forecast plus constraints. It is the constraints, such as forecast and lead-time variability, customer-service levels, costs and capacity that incorporate the risk-management aspect. If the economic order quantity (EOQ) algorithm from such an approach recommends acting on a discount for a bulk purchase, I am inclined to trust the analytics.

Demand Driven InventoryOne cautionary note regarding inventory management comes from my colleague, Bob Davis, and his recent book, Demand-Driven Inventory Optimization, where he makes this rather counterintuitive observation: “A lot of the supply-chain productivity initiatives over the past couple of decades, such as ‘just-in-time (JIT)’ or ‘Kanban,’ have actually accomplished not much more than to simply push the inventory problem either upstream or downstream, like squeezing on a balloon, and in the process making the cost problem, on whole, worse than before ... it does not reduce costs so much as push the costs around. The result was a 5% to 10% increase in the cost of doing business with a production JIT/Kanban system.” There is no free inventory lunch--if you want to push your inventory problem to your suppliers, you will have to pay for the privilege.

Cycle Time

I was unfamiliar with Little’s Law as referenced in the article and had to it look up. Named for John Little, an MIT professor, the law bearing his name comes from probability and queuing theory, expressed algebraically as L = λW, where L is the long-term average number of customers in a stable system, λ represents the long-term average effective arrival rate, and W is the average time spent in the system. For example: Customers arrive at the rate of 10 per hour and stay an average of 0.5 hour. This means we should find the average number of customers in the store at any time to be 5.

Little’s Law has been adapted to apply to the performance of production systems, defining a relationship between Work-In-Process inventory, throughput and flow time of a production system in a steady state, taking the form: Inventory (I) = Throughput (T) x Cycle Time (C).

Since rearrangement of the terms implies that Cycle Time = WIP / Throughput, it would seem that reducing WIP while holding throughput constant will reduce cycle time. When I = C x T is reformulated as C = I / T it makes it appear as if C is the dependent variable, when in practice the cycle or flow time of a machine or production process is a relatively fixed independent variable in the short term that informs and drives inventory requirements. 

Taken out of its retail context, [Little's Law] simply is not applicable to the context of a production line."

Taken out of its retail context, where a long cycle time is meaningful in that it implies longer queues which might discourage new customers from entering the store, it simply is not applicable to the context of a production line. When retail queues get too long, store management opens more registers (i.e. increase throughput), it doesn’t close the doors. But unless there’s so much inventory it blocks the aisles and fire exits, the analogy doesn’t hold in a production environment.

I’m not saying you can’t have too much inventory (I made the case for optimization above), I merely contend that retail queues and factory machine queues are not analogous.

Simply putting a label on a ratio doesn’t make it a meaningful metric, no matter how impressive the objective of reducing cycle time might sound.

However, I do endorse inventory turns as a metric directly related to the cash-to-cash cycle. Similarly, increasing throughput is valuable in its own right as a stand-alone metric; finance says, “Go for it!”  However, if I work to increase T, by necessity the math says I will have to increase inventory to support the improved productivity. And I’ll take that to the bank as a positive outcome any day of the week.